Skip to content
Hallway
0:00 / 0:00

#aixi#kolmogorovcomplexity#solomonoffinduction

Marcus Hutter: Universal Artificial Intelligence, AIXI, and AGI | Lex Fridman Podcast #75

Lex Fridman

5M subscribers • 39 videos on Hallway

YouTube
  • Recommended by 1 person: 1 on Hacker News.
  • Presented by Marcus Hutter, senior research scientist at DeepMind
  • May be out of date. Uploaded 6 years ago, in an area where tools change fast.

Advanced. Deep dive for people who know the field.

126K views6 years ago1 recommendation

AI and machine learningComputer scienceComputation and complexityNeural networks and deep learningMath for machine learning

Lex Fridman interviews Marcus Hutter about universal artificial intelligence, the AIXI model, Kolmogorov complexity, and Solomonoff induction.

Marcus Hutter is a senior research scientist at DeepMind and professor at Australian National University. Throughout his career of research, including with Jürgen Schmidhuber and Shane Legg, he has proposed a lot of interesting ideas in and around the field of artificial general intelligence, including the development of the AIXI model which is a mathematical approach to AGI that incorporates ideas of Kolmogorov complexity, Solomonoff induction, and reinforcement learning. This episode is presented by Cash App. Download it & use code "LexPodcast": Cash App (App Store): apple.co/2sPrUHe Cash App (Google Play): bit.ly/2MlvP5w PODCAST INFO: Podcast website: lexfridman.com/podcast Apple Podcasts: apple.co/2lwqZIr Spotify: open.spotify.com/episode/4BgbFyYFoau4YaeGUu6chb RSS: lexfridman.com/feed/podcast/ Full episodes playlist: youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4 Clips playlist: youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41 EPISODE LINKS: Hutter Prize: prize.hutter1.net Marcus web: hutter1.net Books mentioned: - Universal AI: amzn.to/2waIAuw - AI: A Modern Approach: amzn.to/3camxnY - Reinforcement Learning: amzn.to/2PoANj9 - Theory of Knowledge: amzn.to/3a6Vp7x OUTLINE: - Introduction - Universe as a computer - Occam's razor - Solomonoff induction - Kolmogorov complexity - Cellular automata - What is intelligence? - AIXI - Universal Artificial Intelligence - Where do rewards come from? - Reward function for human existence - Bounded rationality - Approximation in AIXI - Godel machines - Consciousness - AGI community - Book recommendations - Two moments to relive (past and future) CONNECT: - Subscribe to this YouTube channel - Twitter: twitter.com/lexfridman - LinkedIn: linkedin.com/in/lexfridman - Facebook: facebook.com/LexFridmanPage - Instagram: instagram.com/lexfridman - Medium: medium.com/@lexfridman - Support on Patreon: patreon.com/lexfridman

Why it's here

1 person recommends this

What people say. Recommended for its clear explanation of Kolmogorov complexity and the theoretical relationship between intelligence and compression.

jcimson Hacker News • Recommends

On “Machine Learning, Kolmogorov Complexity, and Squishy Bunnies (2019)”

New concepts to me coming from outside the field. Kolmogorov complexity was discussed just recently on Lex Fridman's podcast with Marcus Hutter - link to this video I don't know enough to say how accurate Hutter's explanation is but it made sense to me.
View on Hacker News

WithinReasonon Hacker News • Mentions

On “Hutter Prize”

> Are they? They are actually. It's a kind of revelation when you realise this statement is true. There are people in AI who have thought long about the problem of intelligence and had this realisation. For example: link Then there is the interview with Hutter himself: link to this video You can even semi-formally derive this from Bayes' rule. Let's consider intelligence to be the ability to predict future data from past data. This requires building a model for the past data that can make predictions. So intelligence produces the most likely model M given data D: argmax_M P(M|D) which from Bayes' rule is: argmax_M P(M|D) = P(D|M)*P(M)/P(D) You convert probabilities to information by taking the negati…
View on Hacker News

AIorNoton Hacker News • Mentions

On “ARC-AGI without pretraining”

I was thinking about this lex friedman podcast with Marcus Hutter. Also, Joshua Bach defined intelligence as the ability to accurately model reality.. is lossless compression itself intelligence or a best fit model- is there a difference? link to this video
View on Hacker News